Track the final version of the model

optimizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bb7b05699b568108461a3e8d6d9e24788a0edb5976994b84e1311d25c6baa6fc
+size 1302854789

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:13e32e53df3a892158a15319f4cde205002cc6f0ca783bcf74bf078a1c3b0681
+oid sha256:aba1e23ea7f0ba7f424f3c69b922b42409b699cc42ebee9a33b5dcdc51c6c6e5
 size 651441269

rng_state.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:d9327307ae12330b94a75b0a30d0943899ed791da5c8fd404aec07bf839d1df8
+size 14575

scheduler.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bf501303f049f96d13ee57ddac5410f72a079229358bfc0084c90a64ba044efe
+size 627

trainer_state.json

@@ -0,0 +1,211 @@
+{
+  "best_metric": null,
+  "best_model_checkpoint": null,
+  "epoch": 3.0,
+  "global_step": 9648,
+  "is_hyper_param_search": false,
+  "is_local_process_zero": true,
+  "is_world_process_zero": true,
+  "log_history": [
+    {
+      "epoch": 0.16,
+      "learning_rate": 4.740878938640133e-05,
+      "loss": 0.0457,
+      "step": 500
+    },
+    {
+      "epoch": 0.31,
+      "learning_rate": 4.4817578772802655e-05,
+      "loss": 0.0394,
+      "step": 1000
+    },
+    {
+      "epoch": 0.31,
+      "eval_dev_RMSE": 0.19654961613659372,
+      "eval_dev_loss": 0.038631755858659744,
+      "eval_dev_runtime": 133.0005,
+      "eval_dev_samples_per_second": 92.18,
+      "eval_dev_steps_per_second": 11.526,
+      "step": 1000
+    },
+    {
+      "epoch": 0.47,
+      "learning_rate": 4.222636815920398e-05,
+      "loss": 0.0364,
+      "step": 1500
+    },
+    {
+      "epoch": 0.62,
+      "learning_rate": 3.9635157545605314e-05,
+      "loss": 0.0371,
+      "step": 2000
+    },
+    {
+      "epoch": 0.62,
+      "eval_dev_RMSE": 0.17962921125744657,
+      "eval_dev_loss": 0.03226665407419205,
+      "eval_dev_runtime": 133.0252,
+      "eval_dev_samples_per_second": 92.163,
+      "eval_dev_steps_per_second": 11.524,
+      "step": 2000
+    },
+    {
+      "epoch": 0.78,
+      "learning_rate": 3.7043946932006634e-05,
+      "loss": 0.0347,
+      "step": 2500
+    },
+    {
+      "epoch": 0.93,
+      "learning_rate": 3.445273631840796e-05,
+      "loss": 0.0344,
+      "step": 3000
+    },
+    {
+      "epoch": 0.93,
+      "eval_dev_RMSE": 0.17692162830330602,
+      "eval_dev_loss": 0.031301263719797134,
+      "eval_dev_runtime": 133.0382,
+      "eval_dev_samples_per_second": 92.154,
+      "eval_dev_steps_per_second": 11.523,
+      "step": 3000
+    },
+    {
+      "epoch": 1.09,
+      "learning_rate": 3.1861525704809286e-05,
+      "loss": 0.0271,
+      "step": 3500
+    },
+    {
+      "epoch": 1.24,
+      "learning_rate": 2.9270315091210616e-05,
+      "loss": 0.0218,
+      "step": 4000
+    },
+    {
+      "epoch": 1.24,
+      "eval_dev_RMSE": 0.18239944963895938,
+      "eval_dev_loss": 0.03326955810189247,
+      "eval_dev_runtime": 133.0485,
+      "eval_dev_samples_per_second": 92.147,
+      "eval_dev_steps_per_second": 11.522,
+      "step": 4000
+    },
+    {
+      "epoch": 1.4,
+      "learning_rate": 2.6679104477611942e-05,
+      "loss": 0.0221,
+      "step": 4500
+    },
+    {
+      "epoch": 1.55,
+      "learning_rate": 2.408789386401327e-05,
+      "loss": 0.0229,
+      "step": 5000
+    },
+    {
+      "epoch": 1.55,
+      "eval_dev_RMSE": 0.1816432664222771,
+      "eval_dev_loss": 0.03299427777528763,
+      "eval_dev_runtime": 132.9549,
+      "eval_dev_samples_per_second": 92.212,
+      "eval_dev_steps_per_second": 11.53,
+      "step": 5000
+    },
+    {
+      "epoch": 1.71,
+      "learning_rate": 2.1496683250414595e-05,
+      "loss": 0.0212,
+      "step": 5500
+    },
+    {
+      "epoch": 1.87,
+      "learning_rate": 1.890547263681592e-05,
+      "loss": 0.0211,
+      "step": 6000
+    },
+    {
+      "epoch": 1.87,
+      "eval_dev_RMSE": 0.1761829852854008,
+      "eval_dev_loss": 0.03104044497013092,
+      "eval_dev_runtime": 132.9887,
+      "eval_dev_samples_per_second": 92.188,
+      "eval_dev_steps_per_second": 11.527,
+      "step": 6000
+    },
+    {
+      "epoch": 2.02,
+      "learning_rate": 1.6314262023217248e-05,
+      "loss": 0.0201,
+      "step": 6500
+    },
+    {
+      "epoch": 2.18,
+      "learning_rate": 1.3723051409618576e-05,
+      "loss": 0.0119,
+      "step": 7000
+    },
+    {
+      "epoch": 2.18,
+      "eval_dev_RMSE": 0.17386161956544874,
+      "eval_dev_loss": 0.030227866023778915,
+      "eval_dev_runtime": 132.9554,
+      "eval_dev_samples_per_second": 92.211,
+      "eval_dev_steps_per_second": 11.53,
+      "step": 7000
+    },
+    {
+      "epoch": 2.33,
+      "learning_rate": 1.1131840796019902e-05,
+      "loss": 0.0124,
+      "step": 7500
+    },
+    {
+      "epoch": 2.49,
+      "learning_rate": 8.540630182421228e-06,
+      "loss": 0.0123,
+      "step": 8000
+    },
+    {
+      "epoch": 2.49,
+      "eval_dev_RMSE": 0.18054527644599752,
+      "eval_dev_loss": 0.03259659558534622,
+      "eval_dev_runtime": 132.999,
+      "eval_dev_samples_per_second": 92.181,
+      "eval_dev_steps_per_second": 11.526,
+      "step": 8000
+    },
+    {
+      "epoch": 2.64,
+      "learning_rate": 5.9494195688225545e-06,
+      "loss": 0.0115,
+      "step": 8500
+    },
+    {
+      "epoch": 2.8,
+      "learning_rate": 3.358208955223881e-06,
+      "loss": 0.0117,
+      "step": 9000
+    },
+    {
+      "epoch": 2.8,
+      "eval_dev_RMSE": 0.17937361622906337,
+      "eval_dev_loss": 0.03217489272356033,
+      "eval_dev_runtime": 132.9625,
+      "eval_dev_samples_per_second": 92.206,
+      "eval_dev_steps_per_second": 11.53,
+      "step": 9000
+    },
+    {
+      "epoch": 2.95,
+      "learning_rate": 7.669983416252073e-07,
+      "loss": 0.0116,
+      "step": 9500
+    }
+  ],
+  "max_steps": 9648,
+  "num_train_epochs": 3,
+  "total_flos": 8.121060296152474e+16,
+  "trial_name": null,
+  "trial_params": null
+}

training_args.bin

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:bbc8625719231c0cb60062ac046e1d8ffa4236ab87e5315c7b722394df689979
+size 3515